There has been a convergence of progress in chemistry and biology. Among the important advances resulting from this convergence is the development of methods for generating molecular diversity and for detecting and quantifying small quantities of biological or chemical material. This advance been facilitated by fundamental developments in chemistry, including the development of highly sensitive analytical methods, solid state chemical synthesis, and sensitive and specific biological assay systems.
Analyses of biological interactions and chemical reactions, however, require the use of labels or tags to track and identify the results of such analyses. Typically biological reactions are monitored by radiolabels or direct or indirect enzyme labels. Chemical reactions are also monitored by direct or indirect means, such by linking the reactions to a second reaction in which a colored, fluorescent, chemiluminescent or other such product results. These analytical methods, however, are often time consuming and tedious. There is, thus, a need to develop alternative methods for tracking and identifying analytes in biological interactions and the reactants and products of chemical reactions.
Hybridization Reactions
For example, it is often desirable to detect or quantify very small concentrations of nucleic acids in biological samples. Typically, to perform such measurements, the nucleic acid in the sample i.e., the target nucleic acid! is hybridized to a detection oligonucleotide. In order to obtain a detectable signal proportional to the concentration of the target nucleic acid, either the target nucleic acid in the sample or the detection oligonucleotide is associated with a signal generating reporter element, such as a radioactive atom, a chromogenic or fluorogenic molecule, or an enzyme such as alkaline phosphatase! that catalyzes a reaction that produces a detectable product. Numerous methods are available for detecting and quantifying the signal.
Following hybridization of a detection oligonucleotide with a target, the resulting signal-generating hybrid molecules must be separated from unreacted target and detection oligonucleotides. In order to do so, many of the commonly used assays immobilize the target nucleic acids or detection oligonucleotides on solid supports. Presently available solid supports to which oligonucleotides are linked include nitrocellulose or nylon membranes, activated agarose supports, diazotized cellulose supports and non-porous polystyrene latex solid microspheres. Linkage to a solid support permits fractionation and subsequent identification of the hybridized nucleic acids, since the target nucleic acid may be directly captured by oligonucleotides immobilized on solid supports. More frequently, so-called "sandwich" hybridization systems are used. These systems employ a capture oligonucleotide covalently or otherwise attached to a solid support for capturing detection oligonucleotide-target nucleic acid adducts formed in solution see, e.g., EP 276,302 and Gingeras et al. (1989) Proc. Natl. Acad. Sci. USA 86:1173!. Solid supports with linked oligonucleotides are also used in methods of affinity purification. Following hybridization or affinity purification, however, if identification of the linked molecule or biological material is required, the resulting complexes or hybrids or compounds must be subjected to analyses, such as sequencing.
Immunoassays
Immunoassays also detect or quantify very small concentrations of analytes in biological samples. Many immunoassays utilize solid supports in which antigen or antibody is covalently, non-covalently, or otherwise, such as via a linker, attached to a solid support matrix. The support-bound antigen or antibody is then used as an analyte in the assay. As with nucleic acid analysis, the resulting antibody-antigen complexes or other complexes, depending upon the format used, rely on radiolabels or enzyme labels to detect such complexes.
The use of antibodies to detect and/or quantitate reagents "antigens"! in blood or other body fluids has been widely practiced for many years. Two methods have been most broadly adopted. The first such procedure is the competitive binding assay, in which conditions of limiting antibody are established such that only a fraction usually 30-50%! of a labeled e.g., radioisotope, fluorophore or enzyme! antigen can bind to the amount of antibody in the assay medium. Under those conditions, the addition of unlabeled antigen e.g., in a serum sample to be tested! then competes with the labeled antigen for the limiting antibody binding sites and reduces the amount of labeled antigen that can bind. The degree to which the labeled antigen is able to bind is inversely proportional to the amount of unlabeled antigen present. By separating the antibody-bound from the unbound labeled antigen and then determining the amount of labeled reagent present, the amount of unlabeled antigen in the sample e.g., serum! can be determined.
As an alternative to the competitive binding assay, in the labeled antibody, or "immunometric" assay also known as "sandwich" assay!, an antigen present in the assay fluid is specifically bound to a solid substrate and the amount of antigen bound is then detected by a labeled antibody see, e.g., Miles et al. (1968) Nature 29:186-189; U.S. Pat. No. 3,867,517; U.S. Pat. No. 4,376,110!. Using monoclonal antibodies two-site immunometric assays are available see, e.g., U.S. Pat. No. 4,376,110!. The "sandwich" assay has been broadly adopted in clinical medicine. With increasing interest in "panels" of diagnostic tests, in which a number of different antigens in a fluid are measured, the need to carry out each immunoassay separately becomes a serious limitation of current quantitative assay technology.
Some semi-quantitative detection systems have been developed see, e.g., Buechler et al. (1992) Clin. Chem. 38:1678-1684; and U.S. Pat. No. 5,089,391! for use with immunoassays, but no good technologies yet exist to carefully quantitate a large number of analytes simultaneously see, e.g., Ekins et al. (1990) J. Clin. Immunoassay 13:169-181! or to rapidly and conveniently track, identify and quantitate detected analytes.
Combinatorial Libraries
Drug discovery relies on the ability to identify compounds that interact with a selected target, such as cells, an antibody, receptor, enzyme, transcription factor or the like. Traditional drug discovery involves screening natural products form various sources, or random screening of archived synthetic material. The current trend, however, is to identify such molecules by rational design and/or by screening combinatorial libraries of molecules.
Methods and strategies for generating diverse libraries, primarily peptide- and nucleotide-based oligomer libraries, have been developed using molecular biology methods and/or simultaneous chemical synthesis methodologies see, e.g., Dower et al. (1991) Annu. Rep. Med. Chem. 26:271-280; Fodor et al. (1991) Science 251:767-773; Jung et al. (1992) Angew. Chem. Ind. Ed. Engl. 31:367-383; Zuckerman et al. (1992) Proc. Natl. Acad. Sci. USA 89:4505-4509; Scott et al. (1990) Science 249:386-390; Devlin et al. (1990) Science 249:404-406; Cwirla et al. (1990) Proc. Natl. Acad. Sci. USA 87:6378-6382; and Gallop et al. (1994) J. Medicinal Chemistry 37:1233-1251!. The resulting combinatorial libraries potentially contain millions of pharmaceutically relevant compounds and can be rapidly screened to identify compounds that exhibit a selected activity.
The libraries fall into roughly three categories: fusion-protein-displayed peptide libraries in which random peptides or proteins are presented on the surface of phage particles or proteins expressed from plasmids; support-bound synthetic chemical libraries in which individual compounds or mixtures of compounds are presented on insoluble matrices, such as resin beads see, e.g., Lam et al. (1991) Nature 354:82-84! and cotton supports see, e.g., , Eichler et al. (1993) Biochemistry 32:11035-11041!; and methods in which the compounds are used in solution see, e.g., Houghten et al. (1991) Nature 354:84-86, Houghten et al. (1992) BioTechniques 313:412-421; and Scott et al. (1994) Curr. Opin. Biotechnol. 5:40-48!. There are numerous examples of synthetic peptide and oligonucleotide combinatorial libraries. The present direction in this area is to produce combinatorial libraries that contain non-peptidic small organic molecules. Such libraries are based on either a basis set of monomers that can be combined to form mixtures of diverse organic molecules or that can be combined to form a library based upon a selected pharmacophore monomer.
There are three critical aspects in any combinatorial library: (i) the chemical units of which the library is composed; (ii) generation and categorization of the library, and (iii) identification of library members that interact with the target of interest, and keeping track of intermediary synthesis products and the multitude of molecules in a single vessel.
The generation of such libraries often relies on the use of solid phase synthesis methods, as well as solution phase methods, to produce combinatorial libraries containing tens of millions of compounds that can be screened in diagnostically or pharmacologically relevant in vitro assay systems. In generating large numbers of diverse molecules by stepwise synthesis, the resulting library is a complex mixture in which a particular compound is present at very low concentrations, so that it is difficult or impossible to determine its chemical structure. Various methods exist for ordered synthesis by sequential addition of particular moieties, or by identifying molecules based on spacial positioning on a chip. These methods are cumbersome and ultimately impossible to apply to highly diverse and large libraries.
Thus, an essential element of the combinatorial discovery process, as well as other areas in which molecules are identified and tracked, is the ability to extract the information made available during synthesis of the library or identification of the active components of intermediary structures. While there are several techniques for identification of intermediary products and final products, nanosequencing protocols that provide exact structures are only applicable on mass to naturally occurring linear oligomers such as peptides and amino acids. Mass spectrographic MS! analysis is sufficiently sensitive to determine the exact mass and fragmentation patterns of individual synthesis steps, but complex analytical mass spectrographic strategies are not readily automated nor conveniently performed. Also, mass spectrographic analysis provides at best simple connectivity information, but no stereoisomeric information, and generally cannot discriminate among isomeric monomers. Another problem with mass spectrographic analysis is that it requires pure compounds; structural determinations on complex mixtures is either difficult or impossible. Finally, mass spectrographic analysis is tedious and time consuming. Thus, although there are a multitude of solutions to the generation of libraries, there are no ideal solutions to the problems of identification, tracking and categorization.
Similar problems arise in any screening or analytical process in which large numbers of molecules or biological entities are screened. In any system, once a desired molecule(s) has been isolated, it must be identified. Simple means for identification do not exist. Because of the problems inherent in any labeling procedure, it would be desirable to have alternative means for tracking and quantitating chemical and biological reactions during synthesis and/or screening processes.
Therefore, it is an object herein to provide methods for identification, tracking and categorization of the components of complex mixtures of diverse molecules.